cs4001_2026T2_Q2_NA.pdf
Data Visualization Design · Quiz 2 · May 2026
← Course papers · Start practice / exam
Questions and published explanations below are available without starting a test. Some questions may not have a published solution yet.
Question 2 MCQ · 10.0 marks
[[IMAGE:b997726c54043f90_2_2]]
[[IMAGE:b997726c54043f90_2_3]]
[[IMAGE:b997726c54043f90_3_4]]
[[IMAGE:b997726c54043f90_3_5]]
[[IMAGE:b997726c54043f90_3_6]]





I have answered this question in the offline paper
I have skipped this question
This question does not apply to me
A published solution is not available for this question yet.
Question 3 MCQ · 10.0 marks
[[IMAGE:b997726c54043f90_3_7]]
[[IMAGE:b997726c54043f90_3_8]]
Question 3 (5 marks)
—Fatalities ——Miles per capita
30 12,000
25 10,000
20 8,000
15 6,000
10 4,000
5 2,000
0 0
1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010
The above dual-axis line chart plots two time series against a shared x-axis (year, 1950-2011):
auto fatalities per 100,000 people on a left y-axis, and vehicle miles driven per capita on a right
y-axis, each with its own independently scaled range.
Both variables are quantitative and both are plotted against the same time axis, yet the two
vertical scales are independent of one another. Explain what this does to a reader trying to
compare the two series, and what happens to the apparent relationship between the lines if one
axis range is stretched or compressed while the other is held fixed. Suggest one alternative
encoding that would let the two series be compared honestly.
[[IMAGE:b997726c54043f90_5_11]]
[[IMAGE:b997726c54043f90_5_12]]




I have answered this question in the offline paper
I have skipped this question
This question does not apply to me
A published solution is not available for this question yet.
Question 4 MCQ · 2.0 marks
[[IMAGE:b997726c54043f90_6_13]]
[[IMAGE:b997726c54043f90_6_14]]


Position along a common scale
Position on identical but non-aligned scales
Length
Slope
A published solution is not available for this question yet.
Question 5 MCQ · 2.0 marks
[[IMAGE:b997726c54043f90_7_15]]
[[IMAGE:b997726c54043f90_7_16]]


They have moved from position along a common scale to length, a demotion
of exactly one rank, since each spoke is read as a length outward from the centre
They have moved from length to angle, since a radar chart encodes each
attribute's value as the angle of its spoke
They have moved from position along a common scale to position on non-
aligned scales, with the enclosed polygon area coming to dominate what readers actually judge
The encoding is unchanged — both charts represent each value as a length, so
the difference between them is purely aesthetic
A published solution is not available for this question yet.
Question 6 MCQ · 2.0 marks
[[IMAGE:b997726c54043f90_8_17]]

She has used the mean where the median was appropriate
The sample is too small to support a product decision
Her sample excludes everyone who already cancelled, so it cannot reveal what
drove people away
Feature usage is ordinal and should not be correlated with renewal
A published solution is not available for this question yet.
Question 7 MCQ · 2.0 marks
[[IMAGE:b997726c54043f90_8_18]]

The views differ because different underlying data were used
The fiscal-year view is the more accurate of the two because fiscal years are
the legal reporting period
The calendar-year view supports that claim and the fiscal-year view does not,
even though the data are identical — the choice of aggregation boundary changes the apparent
trend
Aggregation can change how smooth a line looks but cannot change an
apparent trend
A published solution is not available for this question yet.
Question 8 MCQ · 2.0 marks
Which is the only measure of central tendency that can be applied to non-numeric values?
Mean
Medium
Mode
Standard deviation
A published solution is not available for this question yet.
Question 9 MCQ · 2.0 marks
[[IMAGE:b997726c54043f90_9_19]]
[[IMAGE:b997726c54043f90_10_20]]
[[IMAGE:b997726c54043f90_10_21]]



Panel A uses a bar chart, where a line chart would have been the conventional
choice for this kind of engineering data
Panel A omits the undamaged flights and does not place temperature on an
axis, so the relationship Revathi's team was asserting had no way to become visible
Panel A fails to state the team's recommended temperature limit clearly
enough
Panel A contains too many flights for a review board to absorb in a single
sitting
A published solution is not available for this question yet.
Question 10 MSQ · 2.0 marks
[[IMAGE:b997726c54043f90_11_22]]

The question of **where the data came from** is answered by the sourcing line,
and the answer is a party with a direct commercial interest in the result
The question of **whether the data have been peer-reviewed** is answered,
and a presentation at the vendor's own user conference does not constitute peer review
The question of **how the data were collected** is answered, and voluntary
feedback forms mean the respondents selected themselves
The question of **whether the data are taken out of context** cannot be
settled from the sourcing line, since we are not told what else was measured or how many
comparisons were run before 12% was reported
None of the four questions can be answered here, because the sample size is
not stated
All four concerns fall away if the 12% difference is shown to be statistically
significant
A published solution is not available for this question yet.
Question 11 MSQ · 2.0 marks
[[IMAGE:b997726c54043f90_11_23]]
[[IMAGE:b997726c54043f90_12_24]]


Because operators frequently monitor the display peripherally from across the
room, encoding alerts with motion would remain effective where hue and shape discrimination
fall away
If the utility expands from 35 stations to 120, alert detection time will stay
roughly flat in Grid 1 but rise roughly in proportion to station count in Grid 2
The alerts are found pre-attentively in Grid 1 because the orange marker is
unique along the hue channel, and that uniqueness is what Grid 2 destroys
Providing a clear legend for the category colours would restore the rapid
detection of maintenance alerts in case of Grid 2.
Making the alert orange more saturated than every category colour is
sufficient to restore popout, since saturation outranks hue in the channel ranking
Grid 2 can satisfy both requirements by encoding station categories with
shape while reserving colour (hue) exclusively for maintenance alerts.
A published solution is not available for this question yet.
Question 12 MSQ · 2.0 marks
A student reads that homicide accounts for 51.0% of deaths among rap musicians and 51.5%
among hip hop musicians, against 6.0% across all genres, and concludes that rap is an
exceptionally dangerous profession. Which cautions below are sound?
At least half of all rap and hip hop musicians who have ever lived are still alive,
so the observed deaths are a small and unrepresentative slice
The distortion is more severe the younger the genre
Dying violently or unusually may itself contribute to fame, which affects who
enters the sample at all
The figures are expressed as rates per 100,000 musicians, so the comparison
across genres is adjusted
A published solution is not available for this question yet.
Question 13 MSQ · 2.0 marks
The assembly stage of Cleveland's three visual operations draws on the Gestalt laws of perceptual
organisation.
Which of the following pairs a law with its correct statement?
Closure — the mind continues visual, auditory and kinetic patterns
Continuity — the mind perceives a set of individual elements as a single,
recognisable pattern
Common fate — elements that share a direction of movement are perceived
as belonging together
Invariance — the mind recognises simple objects independent of rotation,
translation, scale, deformation and lighting
A published solution is not available for this question yet.
Question 14 MSQ · 2.0 marks
Consider the 1937 Java Railroad timetable graphic shown below. This visualisation, similar to
Marey's graphical train schedule, was discussed in the lecture videos.
[[IMAGE:b997726c54043f90_14_25]]
**Which of the following statements are correct? (Select all that apply.)**

Both time and distance are encoded as position along a common scale
The slope of each line encodes speed
The law of continuity lets a reader trace a single train across the chart
The chart primarily relies on area to communicate differences between trains.
A published solution is not available for this question yet.
Question 15 MSQ · 2.0 marks
Which of the following statements are consistent with how the human visual system detects and
interprets visual information? (Select all that apply.)
Contrast sensitivity is higher in the chrominance channel than the luminance
channel
Surface colour is constructed largely from edge-contrast information
The visual system sees differences rather than absolute values, and is drawn
to edges
Contrast with the background should be maximised when the outlines of
shapes matter
The visual system can misread information based on perceived visual
properties
A published solution is not available for this question yet.
Question 16 MSQ · 2.0 marks
[[IMAGE:b997726c54043f90_15_26]]
[[IMAGE:b997726c54043f90_15_27]]


At each boundary the visual system's activity is increased on the lighter side
and decreased on the darker side, so a uniform fill appears to shade as it nears its neighbours.
Layout B works because the separator removes the shared edge, leaving no
boundary for the effect to form at.
The reports mean the fills are not truly uniform, and re-exporting at higher
colour depth would fix it.
Raising the class count from seven to fifteen would remove the artefact, since
smaller steps create no edge.
A published solution is not available for this question yet.
Question 17 MSQ · 2.0 marks
[[IMAGE:b997726c54043f90_16_28]]

Recruiting respondents through the customer newsletter introduces selection
bias
Asking respondents to recall illness over three years introduces recall bias
Surveying only customers who are still subscribed after two years introduces
publication bias
The company having funded its own evidence page introduces recall bias
Summarising only the eight written-up trials introduces publication bias
The difference in exercise, smoking and check-up habits introduces healthy
user bias
A published solution is not available for this question yet.
Question 18 MSQ · 2.0 marks
Lakshmi's team is rebuilding the state public health dashboard. Its main screen tracks weekly
dengue case counts across fourteen districts, and the audience is district health officers who open
it each Monday morning.
The current version renders each district as a 3D bar chart on a tinted panel, one district per
screen, with a caption block beneath.
Which of the team's proposed changes follow Tufte's principles for graphical excellence?
Printing the week the monsoon broke directly onto the outbreak curve,
instead of noting it in the caption block below
Replacing the fourteen district series with a single state-wide weekly total, so
that officers have less to read
Showing all fourteen districts as small multiples on one screen against a
shared axis, so profiles can be compared across the state at a glance
Replacing the 3D bars with flat bars and removing the tinted panel
background, since neither carries a value that would be lost by erasing it
Encoding weekly cases as the width of a rectangle and weekly growth rate as
its height, so both variables travel in a single mark
Deleting the axis tick labels and district names, since removing ink always
raises the data-ink ratio
A published solution is not available for this question yet.